lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d1502495-3a7a-79a1-093a-3b6dc81d9a32@codeaurora.org>
Date:   Mon, 5 Mar 2018 12:17:26 +0530
From:   Sibi S <sibis@...eaurora.org>
To:     Philipp Zabel <p.zabel@...gutronix.de>, bjorn.andersson@...aro.org
Cc:     linux-remoteproc@...r.kernel.org, linux-kernel@...r.kernel.org,
        devicetree@...r.kernel.org, georgi.djakov@...aro.org,
        jassisinghbrar@...il.com, ohad@...ery.com, mark.rutland@....com,
        robh+dt@...nel.org, kyan@...eaurora.org, sricharan@...eaurora.org,
        akdwived@...eaurora.org, linux-arm-msm@...r.kernel.org
Subject: Re: [PATCH 1/6] reset: qcom: AOSS (Always on subsystem) reset
 controller

Hi Philipp,

Thanks for the review. I will post out the v2 of the patch series.

On 03/02/2018 04:00 PM, Philipp Zabel wrote:
> Hi sibis,
> 
> thank you for the patch. I have a few questions and comments below.
> 
> On Fri, 2018-03-02 at 14:53 +0530, sibis wrote:
>> Add reset controller driver for Qualcomm SDM845 SoC to
>> control reset signals provided by AOSS for Modem, Venus
>> ADSP, GPU, Camera, Wireless, Display subsystem
>>
>> Signed-off-by: sibis <sibis@...eaurora.org>
>> ---
>>   .../devicetree/bindings/reset/qcom,aoss-reset.txt  |  54 +++++++
>>   drivers/reset/Kconfig                              |  10 ++
>>   drivers/reset/Makefile                             |   1 +
>>   drivers/reset/reset-qcom-aoss.c                    | 161 +++++++++++++++++++++
>>   include/dt-bindings/reset/qcom,aoss-sdm845.h       |  17 +++
>>   5 files changed, 243 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/reset/qcom,aoss-reset.txt
>>   create mode 100644 drivers/reset/reset-qcom-aoss.c
>>   create mode 100644 include/dt-bindings/reset/qcom,aoss-sdm845.h
>>
>> diff --git a/Documentation/devicetree/bindings/reset/qcom,aoss-reset.txt b/Documentation/devicetree/bindings/reset/qcom,aoss-reset.txt
>> new file mode 100644
>> index 0000000..5318e14
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/reset/qcom,aoss-reset.txt
>> @@ -0,0 +1,54 @@
>> +Qualcomm AOSS Reset Controller
>> +======================================
>> +
>> +This binding describes a reset-controller found on AOSS (Always on SubSysem)
>> +for Qualcomm SDM845 SoCs.
>> +
>> +Required properties:
>> +- compatible:
>> +	Usage: required
>> +	Value type: <string>
>> +	Definition: must be:
>> +		    "qcom,aoss-reset-sdm845", "syscon"
>> +
>> +- reg:
>> +	Usage: required
>> +	Value type: <prop-encoded-array>
>> +	Definition: must specify the base address and size of the
>> +		    syscon device.
>> +
>> +
>> +- #reset-cells:
>> +	Usage: required
>> +	Value type: <uint>
>> +	Definition: must be 1; cell entry represents the reset index.
>> +
>> +example:
>> +
>> +aoss_reset: qcom,reset-controller@...0100 {
>> +	compatible = "qcom,aoss-reset-sdm845", "syscon";
>> +	reg = <0xc2b0000 0x20004>;
>> +	#reset-cells = <1>;
>> +};
>> +
>> +
>> +Specifying reset lines connected to IP modules
>> +==============================================
>> +
>> +Device nodes that need access to reset lines should
>> +specify them as a reset phandle in their corresponding node as
>> +specified in reset.txt.
>> +
>> +Example:
>> +
>> +	modem-pil@...0000 {
>> +		...
>> +
>> +		resets = <&aoss_reset AOSS_CC_MSS_RESTART>;
>> +		reset-names = "mss_restart";
>> +
>> +		...
>> +        };
>> +
>> +For list of all valid reset indicies see
>> +<dt-bindings/reset/qcom,aoss-sdm845.h>
>> diff --git a/drivers/reset/Kconfig b/drivers/reset/Kconfig
>> index 7fc7769..4b1da86 100644
>> --- a/drivers/reset/Kconfig
>> +++ b/drivers/reset/Kconfig
>> @@ -81,6 +81,16 @@ config RESET_PISTACHIO
>>   	help
>>   	  This enables the reset driver for ImgTec Pistachio SoCs.
>>   
>> +config RESET_QCOM_AOSS
>> +	bool "Qcom AOSS Reset Driver"
>> +	depends on ARCH_QCOM || COMPILE_TEST
> 
> Do all ARCH_QCOM have AOSS? If so, should this be enabled by default?
> 

AOSS reset is currently supported on SDM845 SoCs and not on any of other 
the previously upstreamed Qualcomm SoCs. I will change description to 
convey the same. I needs to explicitly enabled for SDM845 SoCs.

>> +	select MFD_SYSCON
>> +	help
>> +	  This enables the AOSS (Always On SubSystem) reset driver
>> +	  for Qcom SoCs. Say Y if you want to control reset signals
> 
> Is Qcom and Qualcomm interchangeable?
> 

Though it is used interchangeably, it will make more sense if I
change it to Qualcomm here.

>> +	  provided by AOSS for Modem, Venus, ADSP, GPU, Camera,
>> +	  Wireless, Display subsystem. Otherwise, say N.
>> +
>>   config RESET_SIMPLE
>>   	bool "Simple Reset Controller Driver" if COMPILE_TEST
>>   	default ARCH_SOCFPGA || ARCH_STM32 || ARCH_STRATIX10 || ARCH_SUNXI || ARCH_ZX
>> diff --git a/drivers/reset/Makefile b/drivers/reset/Makefile
>> index 132c24f..c30044a 100644
>> --- a/drivers/reset/Makefile
>> +++ b/drivers/reset/Makefile
>> @@ -14,6 +14,7 @@ obj-$(CONFIG_RESET_LPC18XX) += reset-lpc18xx.o
>>   obj-$(CONFIG_RESET_MESON) += reset-meson.o
>>   obj-$(CONFIG_RESET_OXNAS) += reset-oxnas.o
>>   obj-$(CONFIG_RESET_PISTACHIO) += reset-pistachio.o
>> +obj-$(CONFIG_RESET_QCOM_AOSS) += reset-qcom-aoss.o
>>   obj-$(CONFIG_RESET_SIMPLE) += reset-simple.o
>>   obj-$(CONFIG_RESET_SUNXI) += reset-sunxi.o
>>   obj-$(CONFIG_RESET_TI_SCI) += reset-ti-sci.o
>> diff --git a/drivers/reset/reset-qcom-aoss.c b/drivers/reset/reset-qcom-aoss.c
>> new file mode 100644
>> index 0000000..eb8c69b
>> --- /dev/null
>> +++ b/drivers/reset/reset-qcom-aoss.c
>> @@ -0,0 +1,161 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/*
>> + * Copyright (C) 2018 The Linux Foundation. All rights reserved.
>> + */
>> +
>> +#include <linux/mfd/syscon.h>
>> +#include <linux/module.h>
>> +#include <linux/platform_device.h>
>> +#include <linux/reset-controller.h>
>> +#include <linux/regmap.h>
>> +#include <linux/of_device.h>
>> +#include <dt-bindings/reset/qcom,aoss-sdm845.h>
>> +
>> +struct qcom_aoss_reset_map {
>> +	unsigned int reg;
>> +	u8 bit;
>> +};
>> +
>> +struct qcom_aoss_desc {
>> +	const struct regmap_config *config;
>> +	const struct qcom_aoss_reset_map *resets;
>> +	int delay;
>> +	size_t num_resets;
>> +};
>> +
>> +struct qcom_aoss_reset_data {
>> +	struct reset_controller_dev rcdev;
>> +	struct regmap *regmap;
>> +	const struct qcom_aoss_desc *desc;
>> +};
>> +
>> +static const struct regmap_config aoss_sdm845_regmap_config = {
>> +	.name		= "aoss-reset",
>> +	.reg_bits	= 32,
>> +	.reg_stride	= 4,
>> +	.val_bits	= 32,
>> +	.max_register	= 0x20000,
>> +	.fast_io	= true,
>> +};
>> +
>> +static const struct qcom_aoss_reset_map aoss_sdm845_resets[] = {
>> +	[AOSS_CC_MSS_RESTART] = { 0x0, 0 },
>> +	[AOSS_CC_CAMSS_RESTART] = { 0x1000, 0 },
>> +	[AOSS_CC_VENUS_RESTART] = { 0x2000, 0 },
>> +	[AOSS_CC_GPU_RESTART] = { 0x3000, 0 },
>> +	[AOSS_CC_DISPSS_RESTART] = { 0x4000, 0 },
>> +	[AOSS_CC_WCSS_RESTART] = { 0x10000, 0 },
>> +	[AOSS_CC_LPASS_RESTART] = { 0x20000, 0 },
>> +};
>> +
>> +static const struct qcom_aoss_desc aoss_sdm845_desc = {
>> +	.config = &aoss_sdm845_regmap_config,
>> +	.resets = aoss_sdm845_resets,
>> +	/* Wait 6 32kHz sleep cycles for reset */
>> +	.delay = 200,
>> +	.num_resets = ARRAY_SIZE(aoss_sdm845_resets),
>> +};
>> +
>> +static struct qcom_aoss_reset_data *to_qcom_aoss_reset_data(
>> +				struct reset_controller_dev *rcdev)
>> +{
>> +	return container_of(rcdev, struct qcom_aoss_reset_data, rcdev);
>> +}
>> +
>> +static int qcom_aoss_control_assert(struct reset_controller_dev *rcdev,
>> +					unsigned long idx)
>> +{
>> +	struct qcom_aoss_reset_data *data = to_qcom_aoss_reset_data(rcdev);
>> +	const struct qcom_aoss_reset_map *map = &data->desc->resets[idx];
>> +
>> +	if (idx >= rcdev->nr_resets)
>> +		return -EINVAL;
> 
> This check is not necessary. For all drivers that don't set of_xlate,
> the default of_reset_simple_xlate already makes sure no reset_control
> is ever returned with rstc->id >= rcdev->nr_resets.
>

Will remove it

>> +	return regmap_update_bits(data->regmap, map->reg,
>> +					BIT(map->bit), BIT(map->bit));
>> +}
>> +
>> +static int qcom_aoss_control_deassert(struct reset_controller_dev *rcdev,
>> +					unsigned long idx)
>> +{
>> +	struct qcom_aoss_reset_data *data = to_qcom_aoss_reset_data(rcdev);
>> +	const struct qcom_aoss_reset_map *map = &data->desc->resets[idx];
>> +
>> +	if (idx >= rcdev->nr_resets)
>> +		return -EINVAL;
> 
> Same here.
> 

ditto

>> +	return regmap_update_bits(data->regmap, map->reg, BIT(map->bit), 0);
>> +}
>> +
>> +static int qcom_aoss_control_reset(struct reset_controller_dev *rcdev,
>> +					unsigned long idx)
>> +{
>> +	struct qcom_aoss_reset_data *data = to_qcom_aoss_reset_data(rcdev);
>> +	int ret;
>> +
>> +	ret = rcdev->ops->assert(rcdev, idx);
> 
> Don't use the ops function pointers here, just call
> qcom_aoss_control_assert directly.

sure will repace it.

> 
>> +	if (ret)
>> +		return ret;
>> +
>> +	udelay(data->desc->delay);
>> +
>> +	ret = rcdev->ops->deassert(rcdev, idx);
>> +	if (ret)
>> +		return ret;
>> +
>> +	return 0;
> 
> This can be simplified to just
> 
> 	return qcom_aoss_control_deassert(rcdev, idx);
> 

ditto

>> +}
>> +
>> +static const struct reset_control_ops qcom_aoss_reset_ops = {
>> +	.reset = qcom_aoss_control_reset,
>> +	.assert = qcom_aoss_control_assert,
>> +	.deassert = qcom_aoss_control_deassert,
>> +};
>> +
>> +static int qcom_aoss_reset_probe(struct platform_device *pdev)
>> +{
>> +	struct qcom_aoss_reset_data *data;
>> +	struct device *dev = &pdev->dev;
>> +	const struct qcom_aoss_desc *desc;
>> +
>> +	desc = of_device_get_match_data(&pdev->dev);
> 
> You could use dev here instead of pdev->dev.
>

Yeah I will replace all instances of pdev->dev with dev

>> +	if (!desc)
>> +		return -EINVAL;
>> +
>> +	data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL);
>> +	if (!data)
>> +		return -ENOMEM;
>> +
>> +	data->desc = desc;
>> +	data->regmap = syscon_node_to_regmap(dev->of_node);
>> +	if (IS_ERR(data->regmap)) {
>> +		dev_err(dev, "Unable to get aoss-reset regmap");
>> +		return PTR_ERR(data->regmap);
>> +	}
>> +	regmap_attach_dev(dev, data->regmap, desc->config);
>> +
>> +	data->rcdev.owner = THIS_MODULE;
>> +	data->rcdev.ops = &qcom_aoss_reset_ops;
>> +	data->rcdev.nr_resets = desc->num_resets;
>> +	data->rcdev.of_node = pdev->dev.of_node;
> 
> And here.
> 
>> +	return devm_reset_controller_register(&pdev->dev, &data->rcdev);
> 
> And here.
> 
>> +}
>> +
>> +static const struct of_device_id qcom_aoss_reset_of_match[] = {
>> +	{ .compatible = "qcom,aoss-reset-sdm845", .data = &aoss_sdm845_desc},
> 
> Consider adding a space between &aoss_sdm845_desc and } ...
> 

sure will do

>> +	{}
>> +};
>> +
>> +static struct platform_driver qcom_aoss_reset_driver = {
>> +	.probe = qcom_aoss_reset_probe,
>> +	.driver  = {
>> +		.name  = "qcom_aoss_reset",
> 
> ... and removing the superfluous space here.

Will remove the superfluous spacing

> 
>> +		.of_match_table = qcom_aoss_reset_of_match,
>> +	},
>> +};
>> +
>> +builtin_platform_driver(qcom_aoss_reset_driver);
>> +
>> +MODULE_DESCRIPTION("Qualcomm AOSS Reset Driver");
>> +MODULE_LICENSE("GPL v2");
>> diff --git a/include/dt-bindings/reset/qcom,aoss-sdm845.h b/include/dt-bindings/reset/qcom,aoss-sdm845.h
>> new file mode 100644
>> index 0000000..e9b38fc
>> --- /dev/null
>> +++ b/include/dt-bindings/reset/qcom,aoss-sdm845.h
>> @@ -0,0 +1,17 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/*
>> + * Copyright (C) 2018 The Linux Foundation. All rights reserved.
>> + */
>> +
>> +#ifndef _DT_BINDINGS_RESET_AOSS_SDM_845_H
>> +#define _DT_BINDINGS_RESET_AOSS_SDM_845_H
>> +
>> +#define AOSS_CC_MSS_RESTART	0
>> +#define AOSS_CC_CAMSS_RESTART	1
>> +#define AOSS_CC_VENUS_RESTART	2
>> +#define AOSS_CC_GPU_RESTART	3
>> +#define AOSS_CC_DISPSS_RESTART	4
>> +#define AOSS_CC_WCSS_RESTART	5
>> +#define AOSS_CC_LPASS_RESTART	6
>> +
>> +#endif
> 
> regards
> Philipp
> 

-- 
Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc, is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ